Research On Micro-statistical Modelling On The Carbon Fiber Composite With Epoxy-matrix

The dissertation was supported by the project of National Science Foundation "The mechanism of compound sensitive surface and imaging method of stress monitoring for nuclear safety shells". With aiming at the loading-resistance sensing character of carbon fiber layer with resin-matrix, the research with macro-scale and micro-scale on the functional properties and its mechanism of the polymer-matrix carbon fiber layer was carried out. The study includes carbon fiber layer loading sensitivity under different manufacturing technologies, loading sensitivity's varaiance with diffenet carbon fiber surface's modification, the statistical modeling to the carbon fiber layer based on random resistor network. The following conclusions and innovation can be obtained:Firstly, with analysis to the gauge factor of sensing character, carbon fiber layer with manual molding lechnology has high gauge factor and lower stablility to its resistance changing, ther layer with vacuum pressure process has excellent stablility but lower gauge factor.Secondly, The interface properties of carbon fiber within the composite material has a great effect to its functional characteristics. The work is focused on the research of CF electro-mechanical response after its surface modification using acetone and nitric acid oxidation. With characterization by XPS and SEM, there exist the various changes on the CF surface not only the chemical elements or chemical groups. With changing CF's resistance under uniaxial tension, the CF has differerent electro-mechanical response to loading force. Through statistical analysis, the research shows the sensibility of CF to the electro-mechanical effect can be improved by the surface modification with acetone process or nitric acid oxidization.Thirdly, The electro-mechanical response of single carbon fiber under off-axis tension is focused on in this wok in micrometer scale. The resistance of CF was tested under off-axis tension. The experimental result indicates that CF resistance change under different angle's off-axis tension varies nonlinearly. Strain parallel to CF orientation increases resistance of CF, more effective than strain perpendicular to CF does. The angle 62°between CF orientation and loading direction is the critical value for positive and negative electro-mechanical response.The fourth, with modified N×N random resistor network, the modeling to the carbon fiber layer is studied. With concise resistor network,the accuracy and reliability of program to the modified N×N random resistor network is validated.The fifth, touch resistance with two adjacent fibers has different corresponding style to the loading within different loading procedure. Touch resistnce gets lower when loading increases and becomes much higer after the touch spot is broken.The sixth, with modified N×N random resistor network, the loading resistance corresponding character of carbon fiber layer formed by singer filament can be explained soundly. Finally, the statistical modeling to the carbon fiber layer under micrometer-scale is established with 12.2?m average length of fiber touching segment. The resistances of carbon fiber layer with different dimensions are estimated with in accordance to the measuring value. The model can be used to de-couple the loading resistance character of carbon fiber layer.